Evaluation of the recursive model approach for estimating particulate matter infiltration efficiencies using continuous light scattering data

Abstract

Quantifying particulate matter (PM) infiltration efficiencies (F inf ) in individual homes is an important part of PM exposure assessment because individuals spend the majority of time indoors. While F inf of fine PM has most commonly been estimated using tracer species such as sulfur, here we evaluate an alternative that does not require particle collection, weighing and compositional analysis, and can be applied in situations with indoor sources of sulfur, such as environmental tobacco smoke, gas pilot lights, and humidifier use. This alternative method involves applying a recursive mass balance model (recursive model, RM) to continuous indoor and outdoor concentration measurements (e.g., light scattering data from nephelometers). We show that the RM can reliably estimate F inf , a crucial parameter for determining exposure to particles of outdoor origin. The RM F inf estimates showed good agreement with the conventional filter-based sulfur tracer approach. Our simulation results suggest that the RM F inf estimates are minimally impacted by measurement error. In addition, the average light scattering response per unit mass concentration was greater indoors than outdoors; after correcting for differences in light scattering response the median deviation from sulfur F inf was reduced from 15 to 11%. Thus, we have verified the RM applied to light scattering data. We show that the RM method is unable to provide satisfactory estimates of the individual components of F inf (penetration efficiency, air exchange rate, and deposition rate). However, this approach may allow F inf to be estimated in more residences, including those with indoor sources of sulfur. We show that individual homes vary in their infiltration efficiencies, thereby contributing to exposure misclassification in epidemiological studies that assign exposures using ambient monitoring data. This variation across homes indicates the need for home-specific estimation methods, such as the RM or sulfur tracer, instead of techniques that give average estimates of infiltration across homes.

Indoor particulate matter in rural, wood stove heated homes

“…It depends on particle penetration and deposition, as well as the air exchange rate, and in the absence of indoor sources F inf is equal to the indoor-outdoor concentration ratio. Continuous indoor and outdoor PM 2.5 sampling during each of two 48-hour winter visits was exploited to quantify F inf in homes using a well-validated recursive model approach (Allen et al, 2003; Allen et al, 2007). Hourly indoor and outdoor concentration averages were calculated from 1-minute averages, and any hourly average generated from less than thirty minutes of data was excluded.…”

Indoor/Outdoor Relationships, Trends, and Carbonaceous Content of Fine Particulate Matter in Retirement Homes of the Los Angeles Basin

“…To verify these results, the same analysis of the I/O concentration ratios was then repeated by using only nighttime data (from 12:00 midnight to 6:00 a.m.), because at this time, resident activities causing indoor particle generation were expected to be minimal. Conversely, the RM method, which has been validated recently for estimating F inf for PM 2.5 from hourly light scattering data, 30 states that, for a particular species of interest, the average indoor concentration during hour t (C t in ) is equal to the sum of a fraction of the average outdoor concentration during the same hour (C t out ), a fraction of the average indoor concentration remaining from the previous hour (C tϪ1 in ), and the contribution from indoor sources (S t in ):…”

“…The low AERs (0.25-0.33 hr Ϫ1 ) calculated for G1 and G2 are consistent with the structural characteristics of the sampling sites, the low number of open windows and doors, and the presence of central air conditioners. F inf values were determined for OC, EC, PM 2.5 , and PN by using different methods, including the RM developed by Allen et al 29 Here, the RM approach, validated recently for estimating F inf for PM 2.5 from light scattering data, 30 has been applied to OC, EC, and PN data for the first time. F inf estimates were highest for EC (a nonvolatile species mostly found in the 0.1-0.4 m range 38,39 ) and also for OC (probably because of the substantial percentage mass contribution from SOA throughout CHAPS).…”

Modeling residential fine particulate matter infiltration for exposure assessment

“…The RM approach for estimating PM 2.5 F inf has been shown to reliably predict PM 2.5 F inf based on a comparison with the more commonly used sulphur tracer approach (Allen et al, 2007).…”